Among the many challenges facing the development of a molecular-based nanotechnology, the directed assembly of discrete molecular objects, and their controlled integration into macroscopic structures, are fundamental. The selective self-assembly characteristic inherent to certain molecules (for example, the Watson-Crick specific base pairing that occurs between complementary single strands of DNA) is a property that could be exploited to address these challenges. For example, ordered suspensions of gold nanoparticles have been assembled by first functionalizing the nanoparticles with short DNA oligonucleotides and then introducing complementary DNA to tie the individual particles together as the strands hybridise (Mirkin et al, “A DNA-based method for rationally assembling nanoparticles into macroscopic materials”, Nature 382, 607-609 (1996)). In principle, this concept could be employed to tackle the integration of nanoscale elements onto a macroscopic substrate, such as an array of metal electrodes. Providing each electrode is functionalized with anchoring oligonucleotides of a unique sequence, the nanoscale elements will assemble appropriately if they are functionalized with the complementary oligonucleotides.
Indeed, the ability to pattern a surface locally with different molecular monolayers in a well-controlled fashion and with a high spatial resolution has importance for molecular electronics and biotechnology applications (including high density DNA expression analysis and genotyping), as well as for nanoengineering.
However, the selective multiple coating of electrode arrays with appropriate anchor molecules has not yet been demonstrated on structures where the electrode separations are sub-micrometre, limiting the applicability of this procedure for nanoscale assembly.
A number of techniques are available for introducing oligonucleotides or other anchor molecules locally onto a surface, but none of these simultaneously meet the requirements of resolution, speed and the ability to coat different electrodes uniquely. Microdrop dispensing systems provide simple approaches for the controlled multiple coating of an electrode array, but are restricted to a spatial resolution of greater than 10 □m. Micromachining and microcontact printing offer spatial resolutions of several hundred nanometres but lack the ability to effect multiple coating. A much higher resolution (a few tens of nanometres) has been achieved by nanografting, but this technique is slow, lacks a straightforward extension to allow multiple coating, and requires complex and expensive infrastructure. Recently, a variant of the nanografting technique, dip-pen nanolithography, has been reported. This technique uses an AFM (atomic force microscope) tip, coated with the anchor molecules, as a pen to draw onto the surface. The resolution of this coating technique is also on the nanometre scale, but a high level of stability and solubility of the anchor molecules is required.
It is known that monolayers of thiol compounds can be formed on a gold surface by immersing the surface in an aqueous solution containing the thiol molecule of interest. The gold-sulphur bond formed during this spontaneous chemisorption process can undergo reductive cleavage at about −1 V versus a Ag/AgCl reference electrode, leading to electrochemical desorption. Compounds having different functionalities can also form monolayers on gold surfaces and undergo electrochemical desorption. Similarly, monolayers of molecules having other functional groups can assemble on other surfaces, from which electrochemical desorption is also possible.
Electrochemical desorption has been applied by Wilhelm et al in “Patterns of functional proteins formed by local electrochemical desorption of self-assembled monolayers”, Electrochimica Acta., vol 47, No. 1, 2001/September, pages 275-281. The authors form an alkane thiolate monolayer on a gold electrode and use a scanning electrochemical microscope (SECM) to induce local electrochemical desorption at defined regions of the alkane thiolate monolayer by using an ultramicroelectrode (UME) of 10 μm diameter placed about 5 μm above the macroscopic SAM-covered gold electrode. The exposed regions are then able to chemisorb an ω-functionalised thiol or disulphide such as cystamine. Functional proteins can then be coupled to the amino groups present in the modified regions of the monolayer. Because this method relies on desorption from regions of one large electrode using an UME above the electrode it is subject to resolution restrictions and indeed is restricted to a spatial resolution of around 10 μm. It means that it can be difficult to control the system so that desorption is not induced at areas neighbouring the intended area for desorption.
A different technique is described by Tender et al in “Electrochemical patterning of self-assembled monolayers onto microscopic arrays of gold electrodes fabricated by laser ablation”, Langmuir, 1996, 12, 5515-5518. This group describe use of an array of individually-addressable gold microelectrodes. One technique involves adsorption of a monolayer of (1-mercaptoundec-11-yl) hexa(ethylene glycol) (EG6SH) on all electrodes. Electrochemical desorption is then induced from alternating bands of electrodes, by controlling the potential at the electrodes from which desorption is required. Adsorption of a layer of hexadecanethiol (C16SH) is then allowed to adsorb onto the thus-exposed bands. Thus, alternating bands of C16S and EG6S monolayers are obtained. Non specific absorption of BSA-antibody onto the C16S bands is then allowed and BSA binds specifically to the antibody.
It is stated that the extension of electrochemical desorption of SAMs to pattern SAMs of n different ω-substituted alkane thiols onto n individually-addressable microscopic gold elements “should also be straightforward”. However, we are not aware of any further publications by this group along these lines. Furthermore, we believe that the suggested sequential stripping of the EG6S SAM from different elements and exposure to new alkane thiols, using the method described by Tender et al, would result in contamination of previously patterned layers with subsequently introduced alkane thiols.
Therefore it would be desirable to provide a method for the formation of an array of two or more different molecules, the method being capable of giving nanoscale resolution (distance between areas coated with different molecules) and high purity of the individually patterned regions. It would also be desirable to provide such a method which can be carried out conveniently and at high speed.